1

Report on the 2nd Meeting of the IAEA’s EMRAS* I-131 Working Group

Minutes of the meeting held at the

CIEMAT, Madrid, Spain, 31 May to 2 June 2004

The meeting was hosted by the Centro de Investigaciones Energeticas Medioambientales y

Tecnologicas (CIEMAT).

The meeting chaired by Mr. Paweł Krajewski from Central Laboratory for Radiological

Protection, POLAND, was attended by 7 participants from 6 Member States and staff member from the

IAEA

The purpose of the meeting was to:

• discuss the results of the intercomparison exercise for the Plavsk scenario;

• discuss and agree on methods to evaluate and compare the results obtained by different modelers;

• present and discuss other scenarios;

• plan future work activities.

1 BRIEF RECORD OF THE MEETING1.

The meeting was opened by Mr. Tiberio Cabianca, Scientific secretary of the IWG.

Mr. Tiberio made his opening remarks and discussed the Preliminary Agenda.

Mr. David Cancio, Director of CIEMAT, welcomed the participants and expressed his expectations for a

productive week.

The meeting continued with an overview of the IWG activities, made by Mr. Krajewski2. He reminded the

background of environmental assessment modeling, which began in the 1970s, i.e., BIOMOVS (initiated

in 1985), VAMP (1988–1994), and BIOMASS (1996–2001), the EMRAS.

1 The Meeting agenda in given in Appendix A 2 Detailed information on the presentations can be found in the Minutes of EMRAS* I-131 Working Group on the

EMRAS WEB SITE as PDF-documents: http://www-rasanet.iaea.org/projects/emras/emras.asp.

.

2

The major areas of emphasis of IWG activities i.e.:

• identification the most important sources of bias and uncertainty in the model predictions

• improvement of the accuracy of model predictions

• improvement of modeling procedures

During his presentation Mr. Paweł Krajewski described the preparatory activities of the IWG, current

status and suggested discussion on both general and specific EMRAS objectives, as well as its expected

outputs.

The first Questionnaire3 of IWG resulted in identification of 9 modelers who agreed to took part in the

IWG activities (TABLE 1). Also among the several proposed data set for model testing the Scenario

Plavsk has been selected as the most relevant for the first test model comparison.

TABLE 1. Models and participants

Model Participant Name Country, Organization

LIETDOS Ms T. Nedveckaite, Mr V.

Flistovic Lithuania, Institute of Physics

OSCAAR Mr T. Homma Japan, Japan Atomic Energy Research Institute

UniVes Mr B.Kanyár Hungary, University of Veszprém

CLRP Mr P. Krajewski Poland, Central Laboratory for Radiological Protection

ASTRAL Ms C. Duffa France, Institut de Radioprotection et de Sûreté Nucléaire

(IRSN)

ECOSYS-87 Mr M. Ammann Finland, Radiation & Nuclear Safety Authority (STUK)

Plavsk Dose

Calculator Mr S. Simon USA, National Cancer Institute

SPADE V.6 Mr S.Conney,

D. Webbe-Wood UK, Food Standard Agency

CLIMRAD Mr O. Vlasov Russian Federation, Medical Radiological Research Center

At the end of January 2003 the Secenario Plavsk was evaluated and distributed among the participants

together with prediction formularies (in Excel file). The Scenario is available on the EMRAS WEBSITE:

http://www-rasanet.iaea.org/projects/emras/emras.asp.

3 prior to the First Meeting of the IAEA’s EMRAS I-131 Working Group, held at IAEA Headquarters in Vienna, 1-5 September 2003

3

This scenario involves the reconstruction of 131I deposition using 137Cs ground contamination data as

a tracer and the assessment of 131I thyroid burden and committed inhalation and ingestion dose to

thyroid for different age groups in specified locations.

The scenario can be seen as being in two parts:

1. a model test in which predictions of 131I concentration in milk and 131I content in thyroid can be

compared with observed values in the test area; and

2. a model comparison in which predictions of the mean ingestion dose to the thyroid of different

age groups and inhalation dose are compared and analyzed.

Both urban area and rural settlements are included in the test region and pathways contributing to the

dose include terrestrial environment.

The end points considered for model validation:

1. 131I deposition (soil concentration);

2. a time dependent 131I concentration in milk in the period 27 April –30 May 1986;

for 18 milk farm + Town Plavsk situated at different 131I (137Cs) deposition density;

3. 131I thyroid burden for different age groups:

new born, 1-2, 3-7, 8-12, 13-17, adult -for urban population (Plavsk town);

new born, 1-2, 3-7, 8-12, 13-17, adult -for specified rural locations.

The end points considered for model inter-comparison:

1. reconstruction of 131I air concentration for 18 locations and Plavsk town

2. committed doses to thyroid from ingestion

3. (voluntary) inhalation dose contribution to the total dose.

It was agreed to perform preliminary run of prediction and sent results before the 2nd Meeting of the

IAEA’s EMRAS IWG (31 May 2004).

Eight participants sent results, together with questionnaire summarized assumptions and used parameters.

The meeting continued with an presentation of Plavsk Scenario input data and discussion on credibility of

isotopic ratio I-131/Cs-137 and radioiodine plume dynamic over the Plavsk district.

During the afternoon of the first meeting day and following morning session of the second day the several

modelers presented their models and results of predictions. For modelers that could not be attended the

brief presentation was made by Pawel Krajewski. List of these presentations is given in APPENDIX B .

During the afternoon session Irina Zwonowa presented the observed data for the Plavsk region.

These included 131I concentration in milk and 131I content in thyroids of residences in a settlements.

4

Method of thyroid dose estimation based on results of direct measurements of 131I in thyroids of local

people and in local milk was reported. See appendix APPENDIX C.

In the third day of the meeting the comparison of models predictions against measurements for a number

of test points was presented and discussed.

This report is supplemented with full set of predictions and observed data in form of Excel files:

1. SUMMARY OF PREDICTIONS_AIR_DEPOSITION_PASTURE.xls

2. SUMMARY_OF_PREDICTIONS_I_131_in_Milk

3. SUMMARY_OF_PREDICTIONS_I_131_in_Thyroid.xls

4. SUMMARY OF PREDICTIONS_DOSES.xls

These file are available on the EMRAS website

The major topics discussed during the meeting are summarized in the following sections in the next

chapter entitled “EVALUATION OF PRELIMINARY PREDICTIONS FOR PLAVSK SCENARIO”.

5

2 EVALUATION OF PRELIMINARY PREDICTIONS FOR PLAVSK SCENARIO

2.1 Evaluation of 131I deposition (isotopic ratio 131I/137Cs)

The main activity ratio of gamma-emitters deposited after Chernobyl accident to the surface activity of

reference radionuclide 137Cs is presented in Table 4.2.2.2-1 of the Scenario Plavsk description. This ratio

for 131I/137Cs calculated on 10 May 1986 is equal to 3.34 with standard error of 19%. It was evaluated by

Orlov and Pitkevich4,5 and corresponded to wet deposition. There were four points of soil sampling in

May 1986 in Tula region with one point located in the Plavsk district (Town Plavsk with relatively very

high 137Cs deposition of 475 kBq m-2). However, remarkably inhomogeneous 137Cs deposition for whole

Plavsk district that ranged from 20 to 600 kBq⋅m2 indicates that the radioactive fallout can be classified as

mixed (dry&wet) and a regional approach should be applied. For these locations where 137Cs deposition is

less than 150 kBq⋅m2 i.e.: Skorodnoe, Ol'hi, Novo-Nikol'skij, Udarnik, Rossia, Druzhba, Kommunar, Im.

K. Marksa, Vpered k kommunizmu, the semi-empirical relationship between 131I and 137Cs content in soil

might be used:

bI

)(aCs137

131 σσ ×=

where:

I131σ - deposition of 131I on the specified date

Cs137σ - deposition of 137Cs

a, b - constants

This approach has been used by Knatko6 for Bielarus (a= 33.78, b= 0.64) and by Mahonko7 for Kiev and

Tula regions (a= 8.85, b= 0.85)

4 Orlov, M.Yu., Snykov, V.P., Khvalensky, Yu. A., Teslenko, V.P., and Korenev, A.I. 1992. Radioactive contamination of the territory of Belorussia and Russia after the Chernobyl accident. Atomnaya Energiya 72(4):371-376. (In Russian) 5 Pitkevich, V.A., Shershakov, V.M., Duba, V.V., et. al. 1993. Reconsruction of composition of the Chernobyl radionuclide fallout in the territories of Russia. Radiation and Risk 3:62-93. (In Russian) 6 Knatko, V.A., Dorozhok, I. N., 2001. Reconstruction of 131I deposition density in region of Bielarus with estimation of thyroid doses from inhalation of 131I, Rad. Prot. Dosim.,Vol.,93,No.1, pp. 43-48. 7 Mahonko, K. P., Kozłowa, E.G., et. al. 1996. Contamination of regions with 131I after the accident at the Chernobyl NPP and assessment of upper doses from 131I. Atomnaya Energiya 80, 466-471 (in Russian).

6

One can notice that the activity ratios of 131I/ 137Cs evaluated as constant value approach (Orlov 1986) or

as a power function approach (Mahonko) differ less than factor 2.5 and decrease with higher 137Cs content

in soil (see FIG 1).

Most modelers used constant activity ratio8 given in Scenario Plavsk (see FIG 2.) and this contributed to

almost the same values obtained. The differences in uncertainty ranges evaluated by particular models

result from different calculation method of representative mean of 137Cs content in soil for particular

areas of milk farm and personal judgment (see FIG 3).

FIG 1. The activity ratio of 131I/ 137Cs deposited after Chernobyl accident calculated on 10 May 1986 as a function of 137Cs content in soil – different approaches of evaluation

8 in exception of (CLRP)

15.0I )(85.8Cs137

Cs137

131 −σ×=σσ

36.0I )(78.33Cs137

Cs137

131 −σ×=σσ

34.3Cs137

131I =σσ

7

FIG 2. Summary of used activity ratio of 131I/ 137Cs for predictions of 131I deposition in particular areas of milk farm of Plavsk District

FIG 3. Summary of predicted 131deposition in particular areas of milk farm of Plavsk District (131I deposition sorted in ascending order)

8

2.2 131I concentration in grass

In section 4.3.1 of the Scenario Plavsk is stated that high of the grass in 30 April 1986 was low (about 5

cm-10 cm) what gives yield of the grass about 0.05 kg m-2 dry weight. Although, the assumption on low

grass yield resulted in lower interception fraction, however it did not affect remarkably the predictions of 131I concentration in grass (no more than 10 % ) comparing with assumption on full developed grass (yield

0.2 kg m-2 dry weight) as mass interception fraction remained approximately constant.

The ratios of 131I concentration in grass on 1 May 19869 to the 131I deposition calculated for

particular models fit in a range from 0.3 ( UniVes) to 24.2 (CLRP) (see FIG 4) and reflect different

assumptions that were made by particular models with respect to grass dry and wet interception fraction,

time period and fraction of dry and wet deposition and deposition velocity on grass (see TABLE 4). The

interception of airborne and waterborne radionuklides by vegetation especially in a case of unknown

fraction of wet and dry deposition yielded discrepancy of predictions by to one order of magnitude (see

FIG 5).

Weathering half-life of 131I in grass is suggested in Scenario description (Section 5.5 Spectrometric

Measurements of 131I in Milk Samples) where is stated that “the mean effective period of 131I activity

reduction in milk in the Tula region is determined as 4.2 days”). One could evaluate the weathering half-

time of iodine equal to 8.7 days. The values assumed by models ranged from 8.7 to 13 days base on post

Chernobyl knowledge (see TABLE 4) and this parameter had a minor contribution to the discrepancy of

predictions (see FIG 5).

9 Aapproximately the maximum value of I-131 in grass due to deposition on 30 April

9

FIG 4. Ratio of 131I concentration in grass on 1 May 1986 to 131I deposition

FIG 5. Example for predictions of 131I concentration in grass

10

2.3 131I concentration in milk

There are several factors that could have influence on predictions of 131I concentration in milk

2.3.1 Cow consumption rate

The values range of 40-45 kg/day of fresh grass is suggested in Scenario descriptions and were applied by

most of models, however information on gradual transfer of cattle in grazing regime10 after winter period

was not been reflected by measurements.

FIG 6. Example of predictions of 131I concentration in milk where assumption about gradual pasturing pattern failed.

10 beginning from 2 hours in first days, then increasing time of grazing during 7-10 days, it gives approximately daily consumption of 5, 10, 20, 35, 45 kg of fresh grass during following days.

11

2.3.2 Dates of the beginning of pasturing collective and private cattle

According to the interviews of the collective farms directors (Table 4.3.2-2 in Scenario

descriptions) and inhabitants (Table 4.3.2-1 in Scenario Plavsk descriptions ) grazing of collective cattle

began approximately 10 days later than private cows. This phenomena was not reflected by observed data

(see FIG 6). Analysis of available measurements of 131I in milk from collective farms showed that dates of

beginning grazing periods of private cows could be more suitable than dates pointed for collective farms.

Especially the reported dates after 20 May were not asserted by results of measurements (FIG 7, FIG 8).

One could suggest that the date of 12-15 May my be more representative for start grazing in collective

farms as the warm weather with a temperature of about 15 °C steeled on the Plavsk district..

FIG 7. First example of predictions of 131I concentration in milk where assumption on cow pasturing date according to the director of collective farm reports failed

12

FIG 8. Second example of predictions of 131I concentration in milk where assumption on cow pasturing date according to the director of collective farm report failed.

2.3.3 Representatives ness of milk samples

The main uncertainty was in the form of records in working notebooks. The place of sampling was

recorded as the name of a collective farm without a name of a settlement. But there were from 3 to 12

settlements with different level of radioactive contamination in collective farms, so uncertainty of milk

sample for verification of the model calculation should be correlated with scattering of surface

contamination in settlements included in an examined collective farm. A collective farm could have some

milk farms (two or three). A milk sample could be taken from a separate milk farm, from a common

container with mixed milk from all milk farms or from small cans in which inhabitants handed over excess

milk from their private cows for transportation to a milk factory. These details were not pointed in the

working notebooks.

2.3.4 Iodine Metabolic model for cow

The mean milk yield of cows for Plavsk (Tula) region ranges from 5 to 7 L per day what is two

fold less then milk yield for highly productive cow. It implies to evaluate iodine metabolic model in cow

which could assess iodine transfer factor to milk depending on milk yield. The recent data suggest higher

transfer factor in a range of 0.008 to 0.01 [day⋅L-1].

13

2.3.5 Comparison of observed and predicted 131I concentration in milk

The predicted versus observed data for 131I concentration in milk for 18 collective farms and Plavsk town

are presented on FIG 9. The main reasons of discrepancies are discussed in previous sections.

More then 50% of predicted 131I concentrations in milk fit in a range of (3×observed, 1/3×observed) for

most of the models that gives only one order of magnitude uncertainty of predictions. The time when cows

were put on a pasture seems to be the most important factor of miss predictions and needs to be carefully

considered.

FIG 9. Selected predictions against observed data for Plavsk Scenario. Red and blue line indicate range of (3, 1/3) observation interval. In the squeare boxes the percentage of predictions that fall in (3,1/3) interval for particular participant are shown.

14

2.4 131I thyroid content

This task required models to predict the time dependent average 131I content in thyroid for different ages

groups i.e. new born (less then one year old), (2-3 years old), (5-7 years old), (8-12 years old) and adults

(20 years old) in 16 settlements situated on 14 different milk farm areas where direct thyroid

measurements were conducted (see FIG 10). The settlements can be considered as small villages that

belong to the particular milk farm. Participants had to calculate 33 various time series curves together

with 95% uncertainty ranges. There were several topics discussed with respect of model capability to

predict of 131I content in thyroids.

FIG 10. Plavsk district and settlements

15

2.4.1 Milk consumption rate

The milk consumption rate is reported in section 4.4.2 of Scenario descriptions. One could notice that

there is not substantial differences in milk consumption within different ages group in rural and urban

population (see TABLE 2). The variation in milk consumption rates for different age groups contributes to

the predictions uncertainty no more than approximately 30%. The urban population used to drink about

half of milk amount consumed by rural population. The measurements of 131I activity in thyroid present

much higher variability than one could expect from reported data about milk consumption rates. However,

it needs to remark that statistics of measurements of 131I in thyroids is much less than statistics on

consumption habits.

TABLE 2. Milk consumption rates by age for rural and urban inhabitants (L d-1) base on table.

95% confidence interval Age (years) Arithmetic mean Lover Upper Rural population

0.25 - < 0.5 0.38 0.26 0.50 0.5 - < 0.75 0.33 0.24 0.42 0.75 - < 1 0.45 0.35 0.55

1-2 0.56 0.53 0.59 3-7 0.6 0.57 0.63 8-12 0.44 0.38 0.50

13-17, malea 0.59 0.57 0.61 13-17, femalea 0.38 0.36 0.40

> 17, malea 0.71 0.69 0.73 > 17, femalea 0.65 0.64 0.66

Urban population 0.25 - < 0.5 0.08 0.06 0.10 0.5 - < 0.75 0.21 0.17 0.25 0.75 - < 1 0.36 0.32 0.40

1-2 0.4 0.39 0.41 3-7 0.4 0.39 0.41

8-12a 0.3 0.29 0.31 13-17, malea 0.3 0.28 0.32

13-17, femalea 0.22 0.20 0.24 > 17, malea 0.3 0.29 0.31

> 17, femalea 0.25 0.24 0.26

2.4.2 Iodine metabolic model for different age groups

The Excel program base on the iodine metabolism model developed by Johnson (1981) was provided

for these models that did not have possibility of calculation of 131I content in thyroid (see TABLE 3).

Some models used own formulas based on ICRP-56 publication. Although metabolic parameters for particular

age groups differ remarkably i.e. mass of thyroid, thyroid uptake of stable iodine and concentrations in

particular compartments, the 131I content in thyroid is similar in different ages group (see last row in

16

TABLE 3 entitled: “I-131 activity in thyroid due to constant intake rate of 1 Bq”) and individual age does

not contribute more than 20% to the uncertainty of predicted 131I thyroid contents. On the other hand, the

individual characteristics and endemic property might have stronger influence on measured 131I level in

thyroid but it could not be included in Scenario.

TABLE 3. The iodine metabolism model: Johnson, J.R. (1981). Radioiodine Dosimetry. Journal of Radioanalitycal Chemistry, 65, 223-238

Age Groups METABOLIC PARAMETERS

3 month old

1 year old

2 years old

5 years old

10 years old

15 years old woman man

GUT/LUNG fraction 1 1 1 1 1 1 1 1

Body mass Ms [kg] 3.5 7.2 10.9 22 40 58.9 58 70

Thyroid mass Mt [g] 1.63 2.12 2.65 4.39 7.9 12.1 17 20

Daily intake of stable iodine [µg d-1] 10 20.6 31.1 62.8 116 168 166 200

inorganic compartment [µg]: 5 10 16 32 60 85 84 100

thyroid compartment [µg]: 300 300 300 990 3700 8300 10000 12000

organic compartment [µg]: 56 120 170 350 650 940 930 1100

GUT/LUNG iodine uptake rate l1 [d-1] 192 192 192 192 192 192 192 192 Thyroid uptake rate from inorganic comp.

s2 = 65 *( Ms/70) [µg d-1] 3.3 6.7 10.1 20.4 37.1 54.7 53.9 65.0

iodine release rate by thyroid: l3 = s2 /Mt [d-1] 0.0108 0.0223 0.0337 0.0206 0.0100 0.0066 0.0054 0.0054

iodine release from organic comp. to inorganic comp. l4 [d-1] 0.053 0.053 0.053 0.053 0.053 0.053 0.053 0.053

iodine release from inorganic comp. to urine l5 [d-1] 1.92 1.92 1.92 1.92 1.92 1.92 1.92 1.92

iodine release from organic comp. l6 [d-1] 0.005 0.005 0.005 0.005 0.005 0.005 0.005 0.005

I-131 activity in thyroid due to constant intake rate of 1 Bq 2.24 2.02 1.84 2.05 2.23 2.34 2.36 2.36

17

2.4.3 131I thyroid burden from inhalation

Radioiodine uptake due to inhalation of contaminated air was an additional task for Scenario Plavsk as it

required models to reconstruct the 131I airborne concentration. However, the expected 131I activity in

thyroid due to inhalation becomes remarkably low (less than 10%) comparing with expected 131I activity

in thyroid due to ingestion at the date when the thyroids measurements were performed (see FIG 11).

FIG 11. A comparison of the expected activity of 131I in thyroid (dashed lines) due to inhalation to the activity of 131I due to ingestion and inhalation for MRRC (Vlasow) and CLRP (Krajewski) predictions.

2.4.4 131I burden of new born

Predictions of the 131I activity in the thyroid of new born was omitted in this preliminary task because of

lack proper calculation methodology. Nevertheless in the second run of predictions the suggested

methodology should be applied (see APPENDIX E).

18

2.4.5 Comparison of observed and predicted 131I contents in thyroid

A comparison of measured and predicted values for adults inhabitants of Plavsk Town is presented on FIG

12. Measurements of thyroids in Plavsk district were performed about two weeks later after passing of

radioactive cloud i.e. in the period of 13-30 May, but during only one day for each location. The 95%

confidence interval of the mean11 for each age group in particular settlement ranges over two fold of the

mean and is presented on graphs. Verification of the 131I in thyroids variation in time base on

measurements values is difficult because of very short period of measurements. Moreover the source of

contaminated milk (from private or collective farm) is uncertain, nevertheless Scenario description

suggests to evaluate radioiodine intakes base on the predictions of 131I concentration in local milk (for

settlement) rather than milk from collective farms.

The predicted versus observed data for 131I contents in thyroids of individuals from 15 settlements and

Plavsk town are presented on . The main reasons of discrepancies is discussed in previous sections.

Generally, for all models, the predicted activity of 131I in thyroids follow previously predicted 131I

concentration in consumed milk and reflect assumed time when cows had been put on a pasture. About

70% predictions of one modelers that is closest to the observed data (MRRC, Vlasov) fit in a range of

(3×observed, 1/3×observed), therefore the range about one order of magnitude uncertainty was acheved.

For remaining models careful consideration of some initial assumptions might be recommended.

Especially the dates of the beginning of pasturing collective and private cattle needs to be revised. Simple

code errors also were evident for some models.

11 assuming normal distribution

19

FIG 12. Example of predictions of 131I content in thyroid for inhabitants (adults) of Plavsk Town

FIG 13. Selected predictions of 131I thyroid content against observed data for Plavsk Scenario. Red and blue line indicate range of (3, 1/3) observation interval. In the squeare boxes the percentage of predictions that fall in (3,1/3) interval for particular participant are shown.

20

2.5 Reconstruction of 131I concentration in air for Plavsk Scenario

Additional task of Plavsk scenario required participants to reconstruct the 131I concentration in air over

Plavsk districts base on evaluated previously 131I deposition. Consequently the contribution of inhalation

doses to the total doses could be evaluated. Models needed to assess airborne radioiodine partition

(aerosol, elemental, organic), contribution of dry and wet deposition to the total deposition, dry deposition

velocity, washout ratio. Made assumption and used parameters are summarized in TABLE 4.

Predicted values fit in a range factor two for particular milk farm (FIG 14). Most participants assumed an

inhomogeneous concentration of 131I in air that reflected inhomogeneous pattern of 131I deposition. It

implied that average air concentration during cloud passing changed dramatically (by factor 10) over 40

km width Plavsk region. Consequently the parameters that govern deposition level were assumed to be

constant for whole Plavsk district. One model (Ecosys-98/Ammand) assumed a constant 131I

concentration in air over whole Plavsk district. Some questions aroused about reality of this assumptions.

Despite of consistency in predictions, verification of assumptions mentioned above need advanced

analysis using plume dispersion models and more detailed information with respect to meteorological

condition.

FIG 14. Reconstruction of 131I concentration in air for Plavsk Scenario (sorted according 137Cs deposition)

21

2.6 Dose assessment

The final target of the scenario was to predict mean values of the thyroid dose (committed equivalent dose

to thyroid) from ingestion an inhalation for six different age groups i.e. new born, age 1÷3 age, 3÷7, age 8

÷12 and adult for particular milk farm area and Plavsk Town. The example of predicted doses from

ingestion (mainly contaminated milk) are shown on FIG 15. The range of predicted ingestion doses is

about one order of magnitude for each locations that reflects differences in predicted 131I concentration in

milk by particular participants. It needs to remark that to assess ingestion dose the longer period for 131I

concentration in milk had to be produced by model (approx. up to end of July 1986) then it was required

by formularies provided with Scenario Plavsk. The example of predicted doses from inhalation are

presented on (FIG 16) and item normalized to 137Cs deposition on (FIG 17) . The doses calculated for the

same location ranged by factor 7 across different participant and reflect differences in predictions of 131I

concentration in air. Generally, inhalation doses are 10% of the total dose that is in agreement with

evaluation made by Scenario provider. However, variability of inhalation doses in Plavsk district is

determined by range of changeability of 131I concentration in air over 40×60 km area and again it is mater

of discussion.

FIG 15. Example of predictions of thyroid doses from ingestion in Plavsk district (adults) – doses sorted according deposition of 137Cs.

22

FIG 16. Example of predictions of thyroid doses from inhalation in Plavsk district (adults) – doses sorted according deposition of 137Cs

FIG 17. Example of predictions of thyroid doses from inhalation in Plavsk district (adults) – doses sorted and normalized according deposition of 137Cs

23

TABLE 4. Summary of parameters used by participants

1 2 3 4 5 6 7 8 9 PARTICIPANT END POINT

LIE

TD

OS(T.

Nedveckaite)

CL

IMR

AD

(O

.Vlasov)

SPAD

E V

.6 (D.

(Webbe-W

ood)

UniV

es (B

. Kenyar)

EC

OSY

S – 87 (M

. Am

man)

CL

RP

(P. Krajew

ski)

AST

RA

L

(C. D

uffa)

OSC

AR

(T. H

omm

a)

NC

I (S. Sim

on)

131I deposition in Plavsk district 131I deposition calculation in Plavsk district base on Cs-137 deposition data given in Plavsk scenario (say Yes or No)

YES YES YES YES YES YES YES YES YES

Isotopic ratio 131I/137Cs given in scenario 3.34

YES YES YES

Implicitly by reconstruction of air conc. and rate of dep.

YES

8.85 ×( σC

s-137 ) 0.85

YES YES YES

The dynamic of 131I deposition over the Plavsk district:

The date when first plume of airborne 131I arrive at Plavsk district ?

The date when first plume

of airborne 131I depart the Plavsk district ?

30 April 8:00 1 May 8:00

29 April 12:00 30 April 12:00

29 April 18:00 30 April 6:00

Sigmoint distribution with the mean value given by the directors of farms

29 April 13:00 30 April 9:00

29 April 12:00 30 April 9:00

29 April 12:00 30 April 9:00

- -

If you have assume different period of plume arrival for particular milk farms please specify.

NO NO NO Variation

in time NO NO NO

131I concentration in grass grass yield (kg· m-2 fresh weight) at the start of 131I deposition at the end of 131I deposition

0.45 0.45

0.2 0.2

0.085 0.085

0.20 0.55

0.16 0.16

0.08 0.08

0.7 0.7

grass interception fraction (dimensionless

for dry deposition for wet deposition

formula used: 1-exp(-µY), µ =

weathering H ½ [d]

0.36 0.36

1

8.7

Ow

n evaluation

0.7

2.8

13.2

0.7 -

10

3% 4

0.13 0.05

7

10

0.56 5.610-2 11.6

TABLE 4 (cont). Summary of parameters used by participants

24

1 2 3 4 5 6 7 8 9 PARTICIPANT END POINT

LIE

TD

OS(T.

Nedveckaite)

CL

IMR

AD

(O

.Vlasov)

SPAD

E V

.6 (D.

(Webbe-W

ood)

UniV

es (B

. Kenyar)

EC

OSY

S – 87 (M

. Am

man)

CL

RP

(P. Krajew

ski)

AST

RA

L

(C. D

uffa)

OSC

AR

(T. H

omm

a)

NC

I (S. Sim

on)

131 I concentration in milk Consumption rate

Cattle consumption rate of fresh grass [kg·d-1] Cattle inhalation rate [m3· d-1] if applied Soil ingestion [kg ·d-1]

45

not applied

0.25

40

not applied

not

applied

42.5

130

not applied

45

120

not applied

42

not applied

not

applied

Gradual pasturing 5/10/20/

35/45 100

not

applied

50

not applied

not

applied

Iodine cow metabolic model Milk transfer coefficient for 131 I(d·L-1) if applied Max value of 131I secreted in milk after single intake of 1 Bq [Bq·L-1] Equilibrium level of 131I- after chronic intake of 1 Bq per day [Bq·L-1]

0.003 0.01 ? 0.007 0.003 0.008 0.003

131I contents in thyroid (thyroid burden) Milk consumption rates [L⋅ d-1]

Adult Child (8 – 12 years old)

Child (3-7 years old) Babies (1-2 years old)

Newborn

urban/rural

0.667 0.44 0.60 0.56 0.293

urban/rural urban/rural

0.27/0.68 0.3/0.44 0.4/0.6 0.4/0.56

na

urban/rural

0.68 0.45 0.60 0.55 0.40

urban/rural

0.27/0.68 0.3/0.44 0.4/0.6 0.4/0.56

na

urban/rural

0.3/0.6 0.3/0.4 0.4/0.6 0.4/0.6

na

L. vegetables consumption rate(kg⋅ d-1)

if applied Adult

Child (8 – 12 years old) Child (3-7 years old)

Babies (1-2 years old) Newborn

0.03 0.03 0.03 0.03

0

0.05 0.05 0.05 0.02 0.01

Inhalation rate (m3·d-1) Adult

Child (8 – 12 years old) Child (3-7 years old)

Babies (1-2 years old) Newborn

20

13.3 8.8

5.65 2.9

22 15 10 4 2

22.3 15.4 8.6 5.0 na

24 15 10 6 2

Iodine metabolic model: Johnson, J.R. (1981)

(supplied with Plavsk scenario) ICRP-56 publication

Own

YES

YES

YES

Modifications 0.24 0.089

0.21[d] 28 [d]

YES

YES

25

TABLE 4 (cont). Summary of parameters used by participants

1 2 3 4 5 6 7 8 9 PARTICIPANT

END POINT

LIE

TD

OS(T.

Nedveckaite)

CL

IMR

AD

(O

.Vlasov)

SPAD

E V

.6 (D.

(Webbe-W

ood)

UniV

es (B

. Kenyar)

EC

OSY

S – 87 (M

. Am

man)

CL

RP

(P. Krajew

ski)

AST

RA

L

(C. D

uffa)

OSC

AR

(T. H

omm

a)

NC

I (S. Sim

on)

Reconstruction of 131I concentration in air 131 I iodine speciation (%):

reactive gas (I2) particulate organic

30 40 30

-

Not

considered

100

29/30Apr. 25/28 40/55 35/17

dry deposition velocity (m⋅s-1) if considered ?

reactive gas (I2) particulate

organic

9.5× 10-3

1.6×10-3

7.0×10-5

1.0×10-2

Not considered

1.8×10-3

1.5×10-3

1.1×10-2

1.2×10-3

1.1×10-4

washout factor wr (dimensionless, defined as the ratio of the concentration of radionuclide in surface-level precipitation to the concentration in surface level air during the period of rainfall) Wwet deposition (Bq·m-2)= wr× Cconcentration in air(Bq· m-

3)× Rrainfall(m)

reactive gas (I2) particulate

organic

5.0×106 2.5×106

8.0 × 103

2.0×105

3.1×105

2.0×105

2.0×104

2.0×105

500±200

Doses calculation Inhalation dose Time spent indoors [hours per day]

Adult Child (8 – 12 years old) Child (3-7 years old) Babies (1-2 years old) Newborn

House filtration factor (ratio air concentration indoor to air concentration outdoor) Dose conversion factors if different to those given in scenario

Effective 0.7

17 18 18 18 20

Effective 0.7

1.5×10-7

19.3

0.5

Effective 0.59

½?

16

20

Effective 0.6

Ingestion dose Dose conversion factors if

different to those given in scenario

4.5 ×10-

YES

26

3 Concluding remarks

Ten experts in environmental modeling participated in the Plavsk Scenario, including four who had

not previously been involved in the international model testing programs. The contribution of the

participants to the first Scenario Plavsk presumed future success of the launched project.

During the second IWG meeting several aspects of models performance were discussed. The following

problems were identified to be most in need of attention:

1. Cnstant isotopic ratio 131I/137Cs provided by Scenario Plavsk gives fairly good approximation of 131I deposition, however inhomogeneous 137Cs deposition for whole Plavsk district and relatively

short time of rain during the cloud passage (6 hours) indicates that the radioactive fallout can be

classified as mixed (dry&wet) and a regional approach might be applied with more complex

relationship 131I deposition to 137Cs deposition.

2. Model for grass interception fraction in a case of mixed (dry&wet) radioiodine fallout need to be

carefully considered.

3. Uncertainty associated with prediction of 131I concentration in air over the Plavsk district depends

on partition of airborne radioiodine in to different forms (particulate, elemental, organic) during

the passage of radioactive cloud over the Plavsk district as well as meteorological conditions.

4. Inhomogeneous pattern of 131I fallout not necessary reflects changeability of 131I concentration in

air over 40×60 km area. A plume dispersion model are envisaged to verify this assumption.

5. The time when cows from collective farms were put on a pasture seems to be the most important

factor of miss predictions of 131I concentration in milk and consequently ingestion doses. It needs

to be carefully considered. Analysis of measured data indicate on 12-15 May 1986 as probable

period of start grazing in the most of collective farms.

6. There is lack of validated tool to make prediction of 131I thyroid content for new born (age less

then one year). This modeling approach needs to be developed and implemented in

environmental models and verified base on proper Scenario data set.

7. Measurements data provided for Scenario Plavsk i.e. 137Cs deposition, 131I milk concentration, 131I thyroid content were expressed as a arithmetic mean with standard deviation. The 95%

confidence uncertainty ranges of measurements data need to be evaluated base on log-normal

distribution.

8. Assessments of ingestion and inhalation dose that had been performed previously by Scenario

Plavsk provider need to be included as relevant base for models intercomparison.

In general, although IWG was dealing with areas of assessment modeling for which the capabilities are

not yet well established; there is remarkably improvement in models performance comparing with

27

previous radioiodine scenarios. Predictions of the various models were with in a factor of three of the

observations, discrepancies between the estimates of average doses to thyroid produced by most

participant not exceeded a factor of ten. The process of testing independent model calculations against

independent data set also provided useful information to the originators of the test data.

4 Planned schedule of IWG activities:

1. Comments and supplementary files to the “Report on 2nd Meeting of the IAEA’s EMRAS* I-131

Working Group – end of July 2004

I would please especially participants of our meeting to send to Tiberio Cabianka their

presentations during the meeting to be publish on EMRAS Web-Site.

I would also please participants of IWG to check and complete their data especially parameters

presented in Table 4 and write a few statement comments about their predictions.

2. Revised set of independent observation data (milk, thyroid) – in respect with calculation of 95%

uncertainty ranges base on log-normal distribution – 15 August 2004

3. Final calculation of models predictions for Plavsk Scenario and transmission to the IAEA – end of

September 2004 (possible two week shift but no later than absolute dead line on 10 October

2004).

4. Distribution of Warsaw scenario draft – approximately end of October.

5. Comparison of model predictions, final discussion, conclusion, report preparation: 2nd EMRAS

Combined Meetings 8-11 November 2004 in Vienna

6. Presentation of WARSAW Scenario – 2nd EMRAS Combined Meetings 8-11 November 2004

in Vienna,

I look forward to working with you over the next three months on this interesting and important

topic. Please don't hesitate to contact me with any queries or comments.

Paweł Krajewski

IWG Leader

28

APPENDIX A: Provisional Agenda of the Second Meeting of the IAEA’s EMRAS I-131 Working Group, CIEMAT, Madrid, Spain, 31 May to 2 June 2004

Monday, 31 May 2004 10:00 – 10:30 Welcome to the participants T. Cabianca (IAEA)

Administrative arrangements A. Aguero (CIEMAT)

Introduction to the IWG and summary of activities carried out so far

a. Results of the first questionnaire

b. Summary of the first run predictions of PLAVSK

SCENARIO

c. Summary of the current activities (results of the

second questionnaire) discussion

P Krajewski

10:30 – 11:00 Coffee break

Presentation of PLAVSK SCENARIO – Discussion on input scenario data

11:00 – 12:30

a. Credibility of isotopic ratio I-131/Cs-137

b. Contribution of I-131 dry deposition to total

deposition

c. Metabolic model of iodine for cows

d. Default dose conversion factors for different age

group

e. Availability of additional data

I. Zwonowa,

P. Krajewski

12:30 – 14:00 Lunch break

Presentation of model predictions (first run) – Discussion on modellers assumptions, parameter values used, uncertainty of predictions

14:00 – 15:30

f. MRRC O. Vlasov

29

g. ASTRAL C. Duffa

15:30 – 16:00 Coffee break

Presentation of model predictions (first run) cont.

h. Ecosys-87 M. Amman

16:00 – 17:30

i. OSCAAR T. Homma

30

Tuesday, 1 June 2004

Presentation of model predictions (first run) cont. 10:00 – 10:30

j. TAM DYNAMIC B. Kanyar12

10:30 – 11:00 Coffee break

11:00 – 12:30 Presentation of model predictions (first run) cont.

k. LIETDOS-FILSTEG T. Nedveckaite12

l. CLRP P. Krajewski

m. SPADE V$.6 D. Webbe-Wood12

12:30 – 14:00 Lunch break

Presentation of measurements data 14:00 – 15:30 n. Deposition

o. Milk

p. Thyroid

I. Zvonowa

15:30 – 16:00 Coffee break

Comparison of model predictions and observed data 16:00 – 17:30 q. Discussion on identification of major sources of

discrepancy

r. Discussion on uncertainty of predictions

Wednesday, 2 June 2004

Doses calculation (model intercomparison task) 10:00 – 10:30 s. Discussion on applied methodology

t. Discussion on final report of PLAVSK Scenario

u. IWG activities before November meeting (second

run of predictions of PLAVSC scenario)

P Krajewski

12 This presentation will be given by P Krajewski

31

10:30 – 11:00 Coffee break

11:00 – 12:30 Future activities of IWG (selection of next scenario)

Any other Business

Closure of meeting

32

33

APPENDIX B : List of presentations of 2nd meeting participants available as PDF file.

1. CLIMRAD

Mr O. Vlasov. Russian Federation, Medical Radiological Research Cente’r

2. ASTRAL

Ms C. Duffa France, Institut de Radioprotection et de Sûreté Nucléaire (IRSN)

3. OSCAAR Mr T. Homma, Japan,

Japan Atomic Energy Research Institute

4. ECOSYS-87 Mr M. Ammann, Finland,

Radiation & Nuclear Safety Authority (STUK)

5. IWG activities Summary

6. Comments to INPUT DATA of Scenario Plavsk

7. Presentation of results of predictions (firs run)

Mr Paweł Krajewski

IWG leader

APPENDIX C: Report of Irina Zvonova on participation in the iodine working group meeting, Madrid, 31 May - 2 June, 2004

Preliminary results of model calculations by the “Plavsk” scenario were discussed at the iodine

working group meeting in Madrid, 31 May - 2 June, 2004. According to the scenario the modelers

calculated dynamics of 131I concentration in milk and 131I content in thyroid of residents of the Plavsk

district of Tula region of Russia after the Chernobyl accident. Results of calculations were compared

between themselves and with results of direct measurements.

My participation consisted in the presentation of the measurement results for verification the

model calculations. Results of spectrometric measurements of milk samples from 18 collective farms and

from town of Plavsk collected in May 1986 were presented for discussion. From one to 7 measurements

were made in the examined farms from 14 to 30 of May 1986. Results of direct measurements of 131I

34

thyroid content in inhabitants of 16 settlements of the Plavsk district in May 1986 were also presented for

the verification of the calculations. Methods of 131I measurements in milk samples and in human thyroid,

local features of inhabitant’s lifestyle, grazing of cattle, milk consumption and processing were reported

and explained in details. I answered on questions of participants.

Input parameters for calculations and reasons of discrepancy discussed at the meeting. During the

discussion on ratio 131I/137Cs in fallout depending on 137Cs surface contamination I presented additional

data based on literature and unpublished data that demonstrated the higher ratio 131I/137Cs in fallouts with

low 137Cs surface density (dry deposition) in comparison with the value attached to high 137Cs surface

density typical for wet deposition on the territory of the Bryansk-Belarus 137Cs spot.

Dates of beginning pasture period were discussed at the meeting. According to the interviews of

inhabitants in 1987 and interviews of heads of the collective farms in 1994 grazing of collective cattle

began later than private cows. Analysis of available measurements of collective milk showed that dates of

beginning grazing periods of private cows are more suitable for collective cattle than dates pointed for

collective farms.

Uncertainty of information on milk measurements discussed at the meeting. The main uncertainty

was in the form of records in working notebooks. The place of sampling was recorded as the name of a

collective farm without a name of a settlement. But there were from 3 to 12 settlements with different

level of radioactive contamination in collective farms, so uncertainty of milk sample for verification of the

model calculation should be correlated with scattering of surface contamination in settlements included in

an examined collective farm. A collective farm could have some milk farms (two or three). A milk sample

could be taken from a separate milk farm, from a common container with mixed milk from all milk farms

or from small cans in which inhabitants handed over excess milk from their private cows for

transportation to a milk factory. These details were not pointed in the working notebooks.

Information on I-131 measurements in thyroid of inhabitants of the Plavsk district was presented

for comparison with calculated results. Parameters of generalized distribution of I-131 content in thyroids

of residences in a settlement in one date of measurements was presented.

Method of thyroid dose estimation based on results of direct measurements of I-131 in thyroids of

local people and in local milk was reported at the meeting. This method was used for thyroid dose

reconstruction in Russia.

35

36

APPENDIX D: Milk measurements

Irina Zvonova Institute of Radiation Hygiene

Mira St.8, 197101 St. Petersburg, Russia E-mail: irvaz@iz10087.spb.edu

Spectrometric measurements of 131I content in milk samples and other foods in Tula

region were carried out from May 14 till June 12, 1986 on the base of the regional sanitary-and-

epidemiological station. For this purpose there was used a spectrometer on the base of the crystal

of NaI(Tl) ∅40 х 40 mm, used to define the function of the thyroid gland. The detector was put

inside the cylindrical leaden collimator in vertical position. The sample, which volume was 100

millilitres, in a thin glass was positioned on the butt of the detector. The measurement lasted for

100 seconds. Measurements were done in the energetic channel of gamma-radiation of 131I: 150-

450 keV. Minimally detected activity (MDA) of 131I in sample with 100 seconds of measurement

and 95 % of confidence probability was equal to 0.2 kBq/l. Totally there were performed 1744

measurements of milk samples.

When elaborating the measurement results it happened to be necessary to exclude those

samples that arrived from the settlements, for which the soil contamination with caesium-137 has

not been defined yet. Besides, measurements done in June 1986 were not used either, as in this

period the number of measurements with sample’s activity lower than the MDA increased so

much, that it affected the results of averaging-outs. Thus, the operative data range counted 867

spectrometric measurements performed from May 14 till May 30.

All the measurements of 131I concentration in milk got normalized for 1 kBq/m2 of soil

contamination with 137Cs on the territory where the sample has been taken. When a collective

farm was noted as the place of the sample’s origin, the average contamination value for this

collective farm was taken into consideration. Measurements distribution of normalized 131I

concentration in milk for one day of sample selection had an asymmetric nature, which can be

described well with a logarithmically normal distribution. Therefore when calculating the average

values of the relative concentrations for each day of measurements a logarithmically normal

distribution was always used. On fig. 13 the logarithm dynamics is shown for the relative 131I

concentrations in milk in the Tula region, basing on the elaboration results of all the

measurements performed within May 14 and May 30 1986. The average values of logarithms of

37

the relative concentrations for each day of measurements are performed in table 30. The effective

period of milk decontamination from iodine-131 in the Tula region s defined as 4,2 days.

38

Fig.1. Dynamics of 131I concentration in milk samples from the Tula region.

0 50 100 150 200 250 300 350 400 4500

5

10

15

20

25

30

35

40

45

131 I i

n m

ilk o

n 08

.05.

86, k

Bq/

kg

137Cs in soil, kBq/m 2

Fig. 2. Dependence of the referent 131I concentration in milk on soil contamination with 137Cs in Tula

region

Having the received meaning of the effective period, all the milk measurements were recalculated

for a conventional date May 8, 1986. Concentration values reconstructed for this date were called the

reference 131I concentration in milk. All the reference concentrations in milk samples were consequently

12 16 20 24 28 320

1

2

3

4

5

Ln(13

1 Cm/σ

137),

(Bq13

1 I/kg)

/(kB

q137 C

s/m

2 )

T1/2= 4.2 days

Days after fallout.

39

classified by the range of soil contamination with 137Cs in the place of the sample’s origin. The average

values of soil contamination and reference concentration were calculated from these data. Averaging

results are reported in table 31, and on fig. 14 – a graphics of the reference 131I concentration in milk

depending on soil contamination with 137Сs can be seen [20, 21]. Every point on the graphics represents an

averaging of 16 to 69 measurements of milk samples from villages, which contamination with caesium-

137 remains within the indicated range. Most of villages have had a relatively low radioactive

contamination, so in the range of the contamination less than 74 kBq/m2 the graphics’ points are very

close to each other and represent a higher number of measurements (about 60 each). There are fewer

measurements from villages with a relatively high contamination level; they are classified by a wider

range of soil contamination with 137Cs.

The relation between the reference 131I concentration in milk and the territory’s contamination for

rural settlements is described by the equation:

Crm = (4.0 +0,094⋅σ137) ⋅ [1-exp(-0.693 ⋅ (σ137-2.5)/5)], (2)

where σ137 – density of soil contamination with caesium-137, kBq/m2.

Transfer factor for 08.05.86: 0,094 kBq/l of 131I per 1 kBq/m2 of 137Cs in soil

For 30.05.86: 0,0025 - “ -

Reference 131I concentration in milk of a town, Crm(town), is connected with the average

reference iodine-131 concentration in milk of the district that provides this town with milk, rCm, kBq/kg

by the equation:

Crm(town) = (0,75±0,05)⋅ Cr

m, kBq/l, R=0.93. (3)

Iodine-131 measurements in thyroids of inhabitants

The 131I content in the thyroid of people in the districts of the Tula region, which were contaminated by

radioactive fallouts, was being measured in the radioisotope laboratory of Tula regional hospital from

May, 13 till June 6, 1986. The measurements were done on the standard Russian-made equipment for

radio diagnostics GTRM used for the functional diagnostics of the thyroid gland with 131I. Thyroid

measurements were done with collimated scintillation detector with a NaI(Tl) crystal ∅40 x 40 mm in the

energetic channel of 131I: 300-450 keV [21, 29, 30]. The equipment was calibrated according to the

methods of diagnostic observation of patient’s thyroid, using radioactive 131I in solution Na131I, prepared

and certified by the national corporation “Isotope”. When calibrating, the distance between the detector

and the radioactive source was of 13,5 cm, the same as during the diagnostic measurements of patients.

40

People from the contaminated areas got measured in the position with the neck straight against the

collimator. As soon as the detector was deepened inside the collimator, the distance between the neck of a

measured person and the detector was 8 cm. This correction was introduced when estimating 131I activity

in the thyroid by measurement results. The minimal detectable activity of the apparatus is 0.5 kBq with the

measurement time 60 seconds [29, 30].

Totally 643 people were measured from 40 settlements of the Tula region, where 385

persons were from rural settlements and 258 from urban settlements. Most of the measured

people (459 persons) were living in the Plavsk district. The age distribution of measured urban

and rural inhabitants is shown in table 58.

The results of iodine-131 activity measurements in the thyroid of people living in the district of

interest for some measurements dates are reported in table 1.

APPENDIX E: Proposed methodology to calculate 131I activity in the thyroid of newborn

DOSES TO THE EMBRYO, FETUS AND NEWBORN CHILD FOLLOWING INTAKES OF

RADIONUCLIDES AT WORK BY THE MOTHER

J.R. Gill*, J.D. Harrison+ , A.W. Phipps+ and T. J. Silk+

*Health and Safety Executive, Bootle, Merseyside L20 3QZ.

+National Radiological Protection Board, Chilton, Didcot, Oxon OX11 0RQ.

Abstract

Estimates have been made of in utero and postnatal doses to the child after intakes of radionuclides by the

mother during pregnancy and doses from the ingestion of radionuclides in breast milk. Results show that

working conditions for pregnant women potentially exposed to tritiated water (HTO), 90Sr or 131I should

take account of the possibility that doses to the child could be greater than reference adult values.

Preliminary estimates suggest that exposures of women during breastfeeding to HTO, 90Sr, 131I or 137Cs

may result in doses to the child that are similar to reference adult values.

41

Introduction

The European Union Basic Safety Standards Directive(1) requires the working conditions for pregnant

women to be such that the dose to the child will be as low as reasonably achievable and unlikely to exceed

1 mSv before birth. The directive also prohibits the employment of women who are breastfeeding in areas

of work involving a significant risk of bodily radioactive contamination. The relationships between

maternal intakes of radionuclides and doses to the child during pregnancy and breastfeeding are not

straightforward. This paper presents preliminary estimates of doses to the embryo, fetus and newborn

child, following intakes of radionuclides by the mother before conception, during pregnancy and during

breastfeeding. The radionuclides considered here are 3H, 90Sr, 131I, 137Cs, and 239Pu. The aim of this work

is to provide guidance to employers to help identify types of work where these considerations are

important.

Modelling approaches

Limited information is available on changes during pregnancy in factors such as placental transfer of

radionuclides, distribution within the fetus, geometric relationship of the fetus to maternal organs, and

tissue radiosensitivity. The calculation of doses to the embryo and fetus therefore requires a number of

assumptions. The approaches adopted here, summarised below, are as developed for an International

Commission on Radiological Protection (ICRP) report due for publication in 1999.

Dose to the embryo to the end of week 8 of gestation is taken to be the same as the dose to the uterine

wall.

Photon doses to the whole fetus from week 9 to term (38 weeks) from radionuclides in maternal tissues

and the placenta are calculated using mathematical phantoms representing the pregnant female at the end

of the 1st, 2nd & 3rd trimesters(2).

Because of the small mass of fetal tissues, the calculations take account of both self dose and irradiation of

other fetal tissues (so called cross-fire).

For most radionuclides, transfer to the fetus is calculated from the ratio of the concentration in the body of

the fetus (CF) relative to the mother (CM). These CF:CM ratios, based mainly on animal data, are assumed

constant after intake. In some cases (e.g. Pu) different values are specified for different times of intake. A

similar ratio, CPl:CM, is used to calculate activity in the placenta.

The distribution of elements between fetal tissues is based on that defined for the 3 month old infant(3).

42

Where better data are available, it is possible to develop specific dynamic models. Such models have been

developed for iodine and the alkaline earth elements.

ICRP(4) tissue weighting factors are used to calculate an in utero effective dose.

Radionuclides present in the fetus at term are distributed to corresponding tissues of postnatal models for

the newborn infant in order to calculate committed doses.

Methods for the calculation of doses to newborn children from radionuclides transferred in breast milk

have yet to be agreed by ICRP. The provisional approach adopted here takes account of limited human

and animal data. Transfer to milk from maternal blood and tissues is considered on an element-specific

basis. Milk consumption is taken to average 850 ml d-1 throughout breastfeeding. Increased intestinal

absorption in infants is taken into account in estimates of radionuclide absorption from milk.

Biokinetic data

Tritium

Tritiated water (HTO) in the body of the fetus and mother are assumed to be in equilibrium at all times

after intake. The water concentration is about 60% higher in the fetus than in the mother, so a CF:CM ratio

of 1.6 is adopted. The concentration of 3H in the placenta is taken to be the same as in maternal tissues

(CPl:CM = 1). For breast milk it is assumed that the HTO content is in equilibrium with maternal body

water.

Iodine

A dynamic model developed by Berkovski(5) has been used for the transfer of I to the embryo and fetus. The

model includes bidirectional fetomaternal transport and retention in the thyroid and other tissues of the fetus.

For breast milk, based on an estimated average I secretion into milk of 45 µg d-1, it is assumed that 20% of the

activity in maternal blood is transferred to milk.

Caesium

On the basis of the available human data, CF:CM and CPl:CM ratios of 1 are adopted for Cs. For breast milk, it

is assumed that there is an initial transfer to milk of 13% of Cs entering maternal blood and that subsequent

daily transfer is 0.2% of the maternal body content(6).

Strontium

A biokinetic model developed by Fell et al.(7) has been used for the transfer of alkaline earths to the human

fetus. The model takes account of changes during pregnancy in maternal bone turnover and urinary

43

excretion and relates fetal transfer to requirements for skeletal calcification. Daily transfer of Sr to the

fetus is taken to be lower than for Ca by a factor of 0.6. For breast milk, daily secretion of Ca is taken to

average 280 µg and Sr secretion to be 0.6 times that of Ca. The fractional absorption of ingested Sr by the

child is assumed to be 0.8 at birth, 0.6 from 2 weeks to 2 months of age and 0.4 to 6 months.

Plutonium

Based on animal data, the CF:CM ratios adopted are 0.03, 0.1, 0.3 and 1 for intakes prior to pregnancy and

during the 1st, 2nd & 3rd trimesters, respectively. The ratio CPl:CM is taken to be 0.1 for intakes before

conception and 5.0 for intakes during pregnancy. The distribution of Pu in the fetus is taken to be the same

as the short-term distribution in the 3 month old infant(3). The concentration of Pu in milk is taken to be

about 10% of that in maternal blood. Fractional absorption of ingested Pu by the child is assumed to be 5

x 10-3 to 6 months of age.

Results and Discussion

Table 1 gives preliminary estimates of doses following inhalation of radionuclides at the beginning of the

15th week of pregnancy and compares total doses to the child, in utero and after birth, with corresponding

reference doses for workers(3). For 3H, 137Cs and 90Sr, doses for intakes later in pregnancy are lower than

for intakes at 15 weeks, primarily because of the shorter duration of fetal exposure, although this is

counteracted to some extent by increases in postnatal doses. For 131I and 239Pu, doses for intakes later in

pregnancy are higher than intakes at 15 weeks by factors of up to about 10 and 2 - 3, respectively. The

dominating effect for 131I is the increasing I requirement of the fetal thyroid. For 239Pu, doses resulting

from greater transfer to the fetus in late pregnancy (3rd trimester) are largely delivered postnatally as a

result of long-term retention in the child. For each radionuclide, doses to the child following intakes by

the mother prior to conception are substantially lower than for intakes during pregnancy.

The dose estimates in Table 1 show that working conditions for pregnant women exposed to HTO, 90Sr

and 131I should take account of the possibility that doses to the child will be greater than reference adult

values. This will be true of other radionuclides as well, and we are continuing to look for important ones

to which women may be occupationally exposed.

Preliminary calculations of doses to children from the ingestion of radionuclides in breast milk have been

made, assuming maternal intake by inhalation (5 µm particles or vapour; as Table 1) shortly after giving

44

birth. Committed effective doses are estimated to be similar to reference doses to workers for HTO, 137Cs, 90Sr and 131I and a small fraction (< 10-4) of worker dose for 239Pu.

Radionuclide e (in utero) e (total)b e (total)/ e(adult)c

HTO (vapour) 3.2 x 10-11 3.2 x 10-11 1.8 137Cs (Type Fd) 2.8 x 10-9 2.9 x 10-9 0.4 90Sr (Type F) 7.1 x 10-8 7.4 x 10-8 2.5 131I (Type F) 3.4 x 10-9 3.4 x 10-9 0.3 239Pu (Type S) 2.2 x 10-10 7.7 x 10-9 0.1

Table 1: Doses (SvBq-1) to the child following maternal inhalation of a 5µm AMAD a

aerosol or vapour at the start of the 15th week of pregnancy.

Note a: AMAD = activity median aerodynamic diameter.

Note b: e(total) is the sum of dose e(in utero) and committed effective dose from

activity remaining in the newborn infant at birth.

Note c: e(adult) is the reference dose coefficient for workers(3).

Note d: ICRP respiratory tract model inhalation categories: F = fast absorption to

blood; S = slow absorption.

References

1. EURATOM. Council Directive 96/29/Euratom of 13 May 1996 laying down basic safety

standards for the protection of the health of workers and the general public against the dangers

arising from ionizing radiation. Official Journal of the European Communities 39, L159/1114

(1996).

2. Stabin, M.G. et al. Mathematical Models and Specific Absorbed Fractions of Photon Energy in

the Nonpregnant Adult Female and at the End of Each Trimester of Pregnancy. ORNL/TM12907.

ORNL, Oak Ridge, Tn, USA. (1995).

3. ICRP. The ICRP Database of Dose Coefficients: Workers and Members of the Public, v1.0.

Elsevier Science Ltd., Oxford (1998).

4. ICRP. 1990 Recommendations of the International Commission on Radiological Protection.

ICRP Publication 60. Ann. ICRP, 21 (13). (1991).

5. Berkovski, V. Radioiodine biokinetics in the mother and offspring. Part 1. Pregnant woman.

Part 2. Embryo, Fetus. Int. Workshop on Thyroid Cancer, Cambridge, July 1998. (In press).

45

6. Gall, M., Mahler, S. and Wirth, E. Transfer of 137Cs into mother’s milk. J. Environ.

Radioactivity 14, 331-339 (1991).

7. Fell, T.P., Harrison, J.D. and Leggett, R.W. A model for the transfer of calcium and strontium to

the fetus. Radiat. Prot. Dosim. 79, 311-315 (1998).